Hyaluronidases (HAses; E. C. 3.1.25) are a group of neutral- and acid-active enzymes found throughout the animal kingdom in organisms as diverse as microbes (e.g, Streptococcus pyogenes, Treponema palladium, and nematodes), bees, wasps, hornet, spiders, scorpions, fish, snakes, lizards, and mammals. Hyaluronidases degrade hyaluronan (HA; also known as hyaluronic acid) and, to a lesser extent, chondroitin sulfates (for a review, see Kreil et al. 1995 Protein Sci. 4:1666-9). Vertebrate hyaluronidases are separated into two general classes: 1) the neutral hyaluronidases, such as the predominantly sperm-associated protein PH20 (Liu et al. 1996 Proc. Natl. Acad. Sci. USA 2:7832-7; Primakoff et al. 1985 J. Cell Biol. 101:2239-44; Lin et al. 1993 Proc. Natl. Acad. Sci. USA 90: 10071-5); and 2) the acid-active hyaluronidases, which have a distinct pH optimum between pH 3.5 to 4.0 and have been described in extracts of liver (Fiszer-Szafarz et al. 1995 Acta Biochim Pol. 42: 31-3), kidney (Komender et al. 1973 Bull. Acad. Pol. Sci. [Biol.] 21:637-41), lung (Thet et al. 1983 Biochem. Biophys. Res. Commun. 117:71-7), brain (Margolis et al. 1972 J. Neutrochem. 19:2325-32), skin (Cashman et al. 1969 Arch. Biochem. Biophys. 135:387-95), placenta, macrophages, fibroblasts (Lien et al. 1990 Biochim Biophys. Acta 1034:318-25; Ruggiero et al. 1987 J. Dent. Res. 66:1283-7), urine (Fiszer-Szafarz et al. supra) and human plasma (De Salegui et al. 1967 Arch. Biochem. Biophys. 120:60-67). Acid-active hyaluronidase activity has also been described in the sera of mammals, though some species exhibit no detectable activity at all (Fiszer-Szafarz et al. 1990 Biol. Cell 68:95-100; De Salegui et al. 1967 supra).
Hyaluronan, the main substrate for hyaluronidase, is a repeating disaccharide of [GlcNAc.beta.1-4GlcUA.beta.1-3].sub.n that exists in vivo as a high molecular weight linear polysaccharide. Degradation of hyaluronan by hyaluronidase is accomplished by either cleavage at P-N-acetyl-hexosamine-[1-4]-glycosidic bonds or cleavage at .beta.-gluconorate-[1.fwdarw.3]-N-acetylglucosamine bonds.
Hyaluronan is found in mammals predominantly in connective tissues, skin, cartilage, and in synovial fluid. Hyaluronan is also the main constituent of the vitreous of the eye. In connective tissue, the water of hydration associated with hyaluronan creates spaces between tissues, thus creating an environment conducive to cell movement and proliferation. Hyaluronan plays a key role in biological phenomena associated with cell motility including rapid development, regeneration, repair, embryogenesis, embryological development, wound healing, angiogenesis, and tumorigenesis (Toole 1991 Cell Biol. Extracell. Matrix, Hay (ed), Plenum Press, New York, 1384-1386; Bertrand et al. 1992 Int. J. Cancer 52:1-6; Knudson et al, 1993 FASEB J. 7:1233-1241). In addition, hyaluronan levels correlate with tumor aggressiveness (Ozello et al. 1960 Cancer Res. 20:600-604; Takeuchi et al. 1976, Cancer Res. 36:2133-2139; Kimata et al. 1983 Cancer Res. 43:1347-1354).
Hyaluronidase is useful as a therapeutic in the treatment of diseases associated with excess hyaluronan and to enhance circulation of physiological fluids and/or therapeutic agents at the site of administration. For example, hyaluronidase has been used to reduce intraocular pressure in the eyes of glaucoma patients through degradation of hyaluronan within the vitreous humor (U.S. Pat. No. 4,820,516, issued Apr. 11, 1989). Hyaluronidase has also been used in cancer therapy as a "spreading agent" to enhance the activity of chemotherapeutics and/or the accessibility of tumors to chemotherapeutics (Schuller et al., 1991, Proc. Amer. Assoc. Cancer Res. 32:173, abstract no. 1034; Czejka et al., 1990, Pharmazie 45:H.9) and has been used in combination with other chemotherapeutic agents in the treatment of a variety of cancers including urinary bladder cancer (Horn et al., 1985, J. Surg. Oncol., 2:304-307), squamous cell carcinoma (Kohno et al., 94, J. Cancer Res. Oncol., 120:293-297), breast cancer (Beckenlehner et al., 1992, J. Cancer Res. Oncol. 118:591-596), and gastrointestinal cancer (Scheithauer et al., 1988, Anticancer Res. 8:391-396). Administration of hyaluronidase also induces responsiveness of previously chemotherapy-resistant tumors of the pancreas, stomach, colon, ovaries, and breast (Baumgartner et al., 1988, Reg. Cancer Treat. 1:55-58; Zanker et al., 1986, Proc. Amer. Assoc. Cancer Res. 27:390). Serum hyaluronidase prevents growth of tumors transplanted into mice (De Maeyer et al., 1992, Int. J. Cancer 51:657-660), while injection of hyaluronidase inhibits tumor formation caused by exposure to carcinogens (Pawlowsli et al., 1979, Int. J. Cancer 23:105-109; Haberman et al., 1981, Proceedings of the 17th Annual Meeting of the American Society of Clinical Oncology, Washington, D.C., 22:105, abstract no. 415). Intravenous or intramuscular injection of hyaluronidase is effective in the treatment of brain cancer (gliomas) (PCT published application no. WO88/02261, published Apr. 7, 1988).
Hyaluronidase expression, and levels of hyaluron, have been associated with tumor development and progression. Levels of a secreted neutral hyaluronidase activity in carcinomas derived from ovary (Miura et al. 1995 Anal. Biochem. 225:333-40), prostate (Lokeshwar et al. 1996 Cancer Res 56:651-7), brain, melanocyte, and colon (Liu et al. 1996 Proc. Natl. Acad. Sci. USA 93:7832-7837) are higher than in normal tissue. This secreted neutral hyaluronidase activity appears similar or identical to the neutral hyaluronidase activity of the sperm hyaluronidase PH20. In contrast to neutral activity, the acid active serum hyaluronidase activity is significantly decreased in metastatic carcinomas of the lung, breast, and colon (Northrup et al. 1973 Clin. Biochem. 6:220-8; Kolarova et al. 1970 Neoplasma 17:641-8). Further, mice having an allele of the hyal-1 locus that is associated with lower levels of serum hyaluronidase activity exhibit faster rates of growth of transplanted tumors than mice having an hyal-1 allele that is associated with 3-fold higher hyaluronidase activity levels (Fiszer-Szafarz et al. 1989 Somat. Cell. Mol. Genet. 15:79-83; De Maeyer et al. supra).
At present, the only hyaluronidase activity available for clinical use is a hyaluronidase isolated from a testicular extract from cattle (WYDAS.RTM., Wyeth-Ayerst). The bovine extract is not optimum not only because of its non-human source, but also because the extract contains multiple types of hyaluronidases and other as yet undefined components. While the human serum acid-active hyaluronidase activity would be preferred for administration, this hyaluronidase has not been previously isolated or purified. Although previous studies were able to determine that the serum acid-active hyaluronidase activity is not a component of platelets since hyaluronidase activity levels in plasma are comparable to those found in serum (De Salegui et al. 1967 supra), attempts to isolate this acid active hyaluronidase activity from human serum have met with limited success due in part to the stability of the purified activity and the inability to obtain an adequately high specific activity. Immunopurification attempts have been hindered by the inability to identify and isolate antibodies that specifically bind the activity in its native form in plasma. Monoclonal antibodies identified by conventional ELISA techniques bind denatured human plasma hyaluronidase in the ELISA screening assay do not bind the native, folded protein (Harrison et al. 1988 J Reprod Fertil 82:777-85).
Conventional methods for hyaluronidase activity include ELISA-like assays (Delpech et al. 1987 J. Immunol. Methods 104:223-9; Stern et al. 1992 Matrix 12:397-403; Afify et al. 1993 Arch. Biochem. Biophys. 30:434-41; Reissig et al., 1955, J. Biol. Chem. 217:956-966) in which a sample containing hyaluronidase is applied to the well of a microtiter dish having hyaluronan or hyaluronectin non-covalently bound to its surface. HAse present in the sample degrades the HA substrate. The plates are then washed, and HAse activity is detected by examining the plates for remaining HA substrate.
Hyaluronidase activity can also be detected by substrate gel zymography (Guentenhoner et al., 1992, Matrix 12:388-396). In this assay a sample is applied to a SDS-PAGE gel containing hyaluronan and the proteins in the sample separated by electrophoresis. The gel is then incubated in an enzyme assay buffer and subsequently stained to detect the hyaluronan in the gel. Hyaluronidase activity is visualized as a cleared zone in the substrate gel.
These conventional methods for detecting hyaluronidase activity are hampered by both the difficulties in producing a detectably-labeled hyaluronic acid substrate and the technical difficulties in achieving rapid, sensitive, and reproducible detection of hyaluronidase activity. For example, biotin labeling of hyaluronic acid for use in ELISA-like assays has proved reticent to biotinylation since HA contains no free amine groups, the moiety with which activated biotin covalently binds. Prior attempts to solve this problem have focused on use of a biotinylated-HA binding aggrecan peptide from bovine nasal cartilage (Levvy et al. 1966 Method Enzymol. 8:571-584), which requires tedious, time-consuming steps.
Furthermore, conventional hyaluronidase assays use assay plates having HA substrate non-covalently bound to the plate surface, which can lead to both false positive and false negative results. Because the HA substrate is non-covalently bound to the plate surface, the washing step following exposure of the plates to the HAse-containing sample often results in non-specific removal of non-degraded HA substrate on the plate. Thus, the sensitivity of the conventional HAse assay is compromised. HAse activities using gel zymography avoid the problem associated with ELISA-like assays, but are time-consuming (e.g., the test sample and the HA-containing gel are normally incubated for about 18 hr to 24 hr) and can result in artifacts if the gel is improperly loaded with excess protein sample. Moreover, analyses of crude preparations is impossible by gel zymography.
Thus, despite the presence of a desirable acid active plasma hyaluronidase activity, and human blood product companies' economic motivation to obtain any and all useful components from a resource as precious and scarce as human blood, the human plasma fractions containing this acid active hyaluronidase activity are discarded for want of an acceptable method for its isolation and purification.
Given the value of hyaluronidases in chemotherapy, there is a need in the field for a method of identifying and isolating the polypeptide associated with the acid active hyaluronidase activity in human serum.